A reliability analysis method based on an improved dynamic Bayesian network was proposed to address the reliability issues of aircraft braking systems. Firstly, within the framework of dynamic Bayesian networks, combining dynamic logic gates with an improved conditional probability table modeling method, the failure rate of nodes was dynamically adjusted by introducing interference factors to determine the conditional probability table and marginal probability table of each node. Secondly, based on the dynamic operating characteristics of the aircraft braking system, a dynamic Bayesian network model of its normal braking system was constructed, and the reliability functions of each module component were derived based on the dynamic Bayesian network inference algorithm. Finally, the method was validated using simulated data generated by the Monte Carlo method. By comparing the reliability curve changes of each subsystem module before and after introducing interference factors, and combining the reverse inference ability of dynamic Bayesian networks, the posterior probability distribution of each module unit was analyzed to achieve system fault assessment. The experimental results show that the proposed method can effectively characterize the dynamic characteristics of aircraft brake systems, identify potential fault hazards, and provide theoretical support for the diagnosis and maintenance strategy formulation of normal brake systems.
LIU Rong, XU Ji-hui, TIAN Wen-jie, WANG Feng-lan.
Reliability Analysis of Aircraft Braking System Based on Improved Dynamic Bayesian Network[J].
Science Technology and Engineering, 2026, 26(11): 4903-4911 DOI:10.12404/j.issn.1671-1815.2504388
DBN这种以离散化时间片段的方式,通过建立变量节点的条件概率表(conditional probability table,CPT)和边缘概率表(marginal probability table,MPT)来解算不同时间片后验概率的算法,被称为离散时间贝叶斯网络(discrete-time Bayesian network,DTBN)。然而,随着节点数量与时间片数量的增加,高维度概率表的计算量会呈指数级增长;在连续时间条件下,传统的连续时间贝叶斯网络(continuous-time Bayesian network,CTBN)算法通过建立各变量节点的概率密度函数,能够得出任意时刻节点的后验概率,从而大幅减少计算时间。但该算法需针对不同对象建立不同模型,且节点数量增加时建模难度也会随之上升。相比之下,概率表建模方法不仅无需对任务时间进行离散化处理,其得到的概率表还与传统CTBN算法具备相同功能[10]。
LeiXiaoben, GaoGuanghui, LiBingqiang, et al. Research and development of advanced technologies for aircraft braking systems[J]. Journal of Air Force Engineering University, 2022, 23(6): 8-16.
[3]
许丰瑞. 带有液压系统振动约束的飞机防滑刹车控制研究[D]. 长沙: 中南大学, 2023.
[4]
XuFengrui. Research on aircraft anti slip braking control with hydraulic system vibration constraints[D]. Changsha: Central South University, 2023.
YanFang, XiangChenyang, DongLei, et al. Analysis and evaluation of fault propagation behavior of aircraft full electric braking system under DIMA architecture[J]. Journal of Aeronautics and Astronautics, 2021, 42(9): 324674.
[11]
BaiN, LiuX C, LiJ F, et al. An aircraft brake control algorithm with torque compensation basedon RBF neural network[J]. Chinese Journal of Aeronautics, 2024, 37(1): 438-450.
[12]
JoséJ M L, MaraT. Combining model-based and learning-based anomaly detection schemes for increased performance and safety of aircraft braking controllers[J]. Engineering Applications of Artificial Intelligence, 2025, 139: 109551.
LiLei, ZhangXiaoshen, FuYibo, et al. Research on optimization of aircraft brake control law parameters based on improved particle swarm optimization algorithm[J]. Fire and Command Control, 2024, 49(10): 41-48.
XingShangpeng, SuiYang. Dynamic reliability assessment method for cooling water system of nuclear power plant equipment[J]. Science Technology and Engineering, 2022, 22(23): 10092-10101.
LiXiang, HuangHongzhong, HuangPeng, et al. Reliability analysis and fault diagnosis of power system based on dynamic Bayesian network[J]. Journal of University of Electronic Science and Technology of China, 2021, 50(4): 603-608.
WeiWentao, LiMeifu, ZhuDahuan, et al. Research on reliability evaluation method of nuclear energy system based on dynamic Bayesian network[J]. Nuclear Power Engineering, 2023, 44 (S2): 109-114.
QiLin, HuaiYongcheng, ChenXiaolin, et al. Airport resilience evaluation method based on dynamic Bayesian network[J]. Science Technology and Engineering, 2024, 24(35): 15276-15284.
[23]
TaoH H, JiaP, WangX Y, et al. Reliability analysis of subsea control module based on dynamic Bayesian network and digital twin[J]. Reliability Engineering & System Safety, 2024, 248: 110153.
[24]
CaoE Z, LongZ D, MaY, et al. A DBN-based reliability analysis method for ship standby system considering time redundancy[J]. Ocean Engineering, 2025, 333: 121528.
XiaZhenghong, HeHu, Wujiang, et al. Intelligent evaluation algorithm for the degree of damage and replacement requirements of aircraft brake disc brake pads[J]. Chinese Journal of Safety Sciences, 2024 (10): 94-100.
ZhaoXinyu, HeLang, WuYangmin, et al. Analysis of common malfunctions in aircraft hydraulic systems and discussion on maintenance strategies[J]. China Plant Engineering, 2024(22): 192-193.
WangDaqing, WangXiaoli, LiangPing, et al. Reliability assessment of oil transfer station process system based on T-S fault tree and Bayesian network[J]. Science Technology and Engineering, 2025, 25(22): 9621-9630.
JiaChao, ChaiXiaoliang, WuYouxun. Reliability improvement of aero engine based on key parameter reconstruction[J]. Science Technology and Engineering, 2025, 25(23): 10051-10056.